Research On Fault Diagnosis Technology Based On The Control System Of Folding Aerial Vehicle

With the increasing complexity and large-scale of modern control systems, reliability requirements of the various components in the system are also increasing, and the possibility of failure that occurs when the system works is also increased. In the increasingly complicated large-scale automatic control production process, how to ensure the safety in production and environmental protection is a problem that can’t be ignored. For large complicated control system, in the event of failure will seriously affect the person’s life and property safety,and bring immeasurable loss. Therefore, in order to guarantee the actual system‘s reliability, maintainability and safety, we urgently need to establish a monitoring system to supervise the whole control system operation state,detecting the system’s real-time change and fault information. take effective measures to prevent the occurrence of disastrous accident. The development fault detection and diagnosis(FDD) technology is in order to meet the needs of "monitoring system".However, as the “brain” of the high altitude vehicle, its control system is a closed-loop nonlinear, multi-input and multi-output system, these characteristics determines the control system will be very complexity, uncertainty, nonlinearity and limited by resources, this lead to the little prior knowledge, little robustness as well as misdiagnosis and other difficulties in the control system fault diagnosis. What’ more, in previous studies, researchers always deal with the overhead working truck operation arm as a rigid manipulator, without considering the effect of the flexible deformation, but this is an unavoidable problem in the control system fault diagnosis.In this paper, taking account of the coupling effects between the rigid body motion and elastic deformation of the flexible manipulator, firstly we deduced the boom’s nonlinear dynamical equations of the folding altitude vehicles based on the Lagrange principle. Because of the inaccuracy of the modeling and measurement, as well as the external disturbance and load changes, it’s very difficult to get the complete and accurate motion model of the boom system of high altitude vehicle. It will cause some positioning error of the operating arm if we control with general control methods. In order to solve the problem, in this paper, we used hybrid control method to design the control system of the boom system of high altitude vehicle, so that the flexible operating arm system can track the desired trajectory quickly and accurately even in the condition that load changes and other uncertainties. On the basis of the analytical model of the control system, using state estimation method based on observer for fault diagnosis. Because the characters of nonlinear, uncertainty, strong coupling and resource limitation, traditional observer is unable to track the status of high altitude vehicle’s control system accurately, and it will reduce the robustness and accuracy in the fault diagnosis. Support vector machine based on the structural risk minimization principle, can achieve good effect in dealing with nonlinear and high dimension problem of small sample, it is the most popular learning algorithm of the statistical theory in recent years. Least squares support vector machine(LS-SVM) used least squares linear algorithm as the loss function, instead of the quadratic programming method used in traditional support vector machine(SVM), can be very good for function estimation and classification, and the learning speed of SVM is improved. LS-SVM can solve the problem of large scale, its promotion is also very good, but it will reduce the robustness of the support vector machine’s solution. In this paper we used weighted LS-SVM for the design of the observer, make sure the observer can track the state of the system in complex environment effectively through the adjustment of weights, then we can get the residuals between the state estimation value of the observer and the measured value of the control system, extract the fault feature by the analysis of the residual, thus we can diagnose the fault of high altitude vehicle’s control system accurately and quickly,...